fluent

This MCP server provides a smart navigation layer for ANSYS Fluent documentation. It exposes a set of tools (notably search_help, list_topics, and get_manual_link) that allow an AI assistant or client to locate official Fluent documentation URLs, surface relevant manual sections, and guide users to the right resources. When integrated with an MCP client, you can query for Fluent topics (for example flamelet model, read case, turbulence) and receive direct links to ANSYS Help pages along with suggested manuals. The server acts as a lightweight URL router and reference provider that fetches official docs rather than storing static content, enabling up-to-date guidance with minimal maintenance.

To use it, connect your MCP client (such as Claude, Cursor, VS Code extensions, or other compatible tools) to the fluent-mcp-server via the uv command shown in installation steps. Then call the available tools: search_help(query) to retrieve relevant Help URLs and suggested sections; list_topics() to view the 35+ pre-mapped topics; and get_manual_link(manual, section) to obtain a direct link to a specific Fluent manual or section. The workflow typically involves asking a question about Fluent, receiving URLs, and then a subsequent fetch of the actual documentation content for precise answers.

Prerequisites:

• Python 3.10 or higher
• uv (uvx) package manager
• Git (for cloning the repository)

Installation steps:

1. Clone the MCP server repository: git clone https://github.com/your-org/fluent-mcp-server.git cd fluent-mcp-server

2. Install in development mode using uv: uv pip install -e ".[dev]"

3. Run tests (optional but recommended during development): uv run pytest

4. Run the MCP server: uv run fluent-mcp-server

Notes:

• The server is Python-based and expects uv to manage the runtime. If you prefer not to use development mode, install and run the package as you would with a typical Python package, ensuring the entry point is accessible via the uv tool.
• The local development configuration (in the project root) demonstrates how to specify the server command and arguments for an MCP client.
• For production deployments, you can containerize or adapt the uv invocation to your environment while preserving the command structure: {"command": "uv", "args": ["run", "fluent-mcp-server"]}.

Tips and common issues:

• Ensure Python 3.10+ is active in your environment when running the server.
• Use uv to install and run the server to leverage its built-in project reloader and task management.
• If you see module import errors, verify that the repository was installed in editable mode (pip install -e ".[dev]") so local code changes are reflected.
• The server relies on up-to-date Fluent documentation; if you fork or mirror the repo, keep a mechanism to keep dependencies synchronized.
• Environment variables are optional here; you can extend the mcp_config with VAR_NAME placeholders if your deployment adds runtime configuration (e.g., API endpoints, logging levels).
• For debugging, run uv run fluent-mcp-server in a verbose or debug mode if your uv version supports it and consult the project’s tests in tests/ for expected behaviors.